Beginners Guide to Writing Bash Shell Scripts

Creating and Executing Bash Shell Scripts

Many simple, common system administration tasks are accomplished using command-line tools. Tasks with greater complexity often require chaining together multiple commands that pass results between them. Using the Bash shell environment and scripting features, Linux commands are combined into shell scripts to easily solve repetitive and difficult real-world problems. In its simplest form, a Bash shell script is an executable file that contains a list of commands, and possibly with programming logic to control decision-making in the overall task. When well-written, a shell script is a powerful command-line tool on its own, and can be leveraged by other scripts.

Shell scripting proficiency is essential to successful system administration in any operational environment. Working knowledge of shell scripting is crucial in enterprise environments, where script use can improve the efficiency and accuracy of routine task completion. You can create a Bash shell script by opening a new empty file in a text editor. While you can use any text editor, advanced editors, such as vim or emacs, understand Bash shell syntax and can provide color-coded highlighting. This highlighting helps identify common errors such as improper syntax, unpaired quotes, unclosed parenthesis, braces, and brackets, and much more.

Specifying the Command Interpreter

The first line of a script begins with the notation ‘#!’, commonly referred to as sh-bang or shebang, from the names of those two characters, sharp and bang. This specific two-byte magic number notation indicates an interpretive script; syntax that follows the notation is the fully qualified filename for the correct command interpreter needed to process this script’s lines. To understand how magic numbers indicate file types in Linux, see the file(1) and magic(5) man pages. For script files using Bash scripting syntax, the first line of a shell script begins as follows:


Executing a Bash Shell Script

A completed shell script must be executable to run as an ordinary command. Use the chmod command to add execute permission, possibly in conjunction with the chown command to change the file ownership of the script. Grant execute permission only for intended users of the script.

If you place the script in one of the directories listed in the shell’s PATH environmental variable, then you can invoke the shell script using the file name alone as with any other command. The shell uses the first command it finds with that file name; avoid using existing command names for your shell script file name. Alternatively, you can invoke a shell script by entering a path name to the script on the command line. The which command, followed by the file name of the executable script, displays the path name to the command that will be executed.

[[email protected] ~]$ which hello

[[email protected] ~]$ echo $PATH

Quoting Special Characters

A number of characters and words have special meaning to the Bash shell. However, occasionally you will want to use these characters for their literal values, rather than for their special meanings. To do this, use one of three tools to remove (or escape) the special meaning: the backslash (\), single quotes (’’), or double quotes ("").

The backslash escape character removes the special meaning of the single character immediately following it. For example, to display the literal string # not a comment with the echo command, the # sign must not be interpreted by Bash as having special meaning. Place the backslash character in front of the # sign.

[[email protected] ~]$ echo # not a comment

[[email protected] ~]$ echo \# not a comment
# not a comment

When you need to escape more than one character in a text string, either use the escape character multiple times or employ single quotes (’’). Single quotes preserve the literal meaning of all characters they enclose. Observe the escape character and single quotes in action:

[[email protected] ~]$ echo # not a comment #

[us[email protected] ~]$ echo \# not a comment #
# not a comment
[[email protected] ~]$ echo \# not a comment \#
# not a comment #
[[email protected] ~]$ echo '# not a comment #'
# not a comment #

Use double quotation marks to suppress globbing and shell expansion, but still allow command and variable substitution. Variable substitution is conceptually identical to command substitution, but may use optional brace syntax. Observe the examples of various forms of quotation mark use below.

Use single quotation marks to interpret all text literally. Besides suppressing globbing and shell expansion, quotations direct the shell to additionally suppress command and variable substitution. The question mark (?) is a meta-character that also needs protection from expansion.

[[email protected] ~]$ var=$(hostname -s); echo $var
[[email protected] ~]$ echo "***** hostname is ${var} *****"
***** hostname is host *****
[[email protected] ~]$ echo Your username variable is \$USER.
Your username variable is $USER.
[[email protected] ~]$ echo "Will variable $var evaluate to $(hostname -s)?"
Will variable host evaluate to host?
[[email protected] ~]$ echo 'Will variable $var evaluate to $(hostname -s)?'
Will variable $var evaluate to $(hostname -s)?
[[email protected] ~]$ echo "\"Hello, world\""
"Hello, world"
[[email protected] ~]$ echo '"Hello, world"'
"Hello, world"

Providing Output From a Shell Script

The echo command displays arbitrary text by passing the text as an argument to the command. By default, the text displays on standard output (STDOUT), but it can also be directed to standard error (STDERR) using output redirection. In the following simple Bash script, the echo command displays the message “Hello, world” to STDOUT.

[[email protected] ~]$ cat ~/bin/hello

echo "Hello, world"
[[email protected] ~]$ hello
Hello, world

The echo command is widely used in shell scripts to display informational or error messages. These messages can be a helpful indicator of the progress of a script and can be directed either to standard output, standard error, or be redirected to a log file for archiving. When displaying error messages, it is good practice to direct them to STDERR to make it easier to differentiate error messages from normal status messages.

[[email protected] ~]$ cat ~/bin/hello

echo "Hello, world"
echo "ERROR: Houston, we have a problem." >&2
[[email protected] ~]$ hello 2> hello.log
Hello, world

[[email protected] ~]$ cat hello.log
ERROR: Houston, we have a problem.

The echo command can also be very helpful when trying to debug a problematic shell script. The addition of echo statements to the portion of the script that is not behaving as expected can help clarify the commands being executed, as well as the values of variables being invoked.